1
1.0 INTRODUCTION
The structure of forests and woodlands is characterized by a heterogeneous mixture of patches at different stages of recovery from disturbances and gap replacements (Shugart et al.
2010). According to Zenner (2004) structural components include attributes such as tree species identities, sizes, canopy, dead trees, coarse debris and the interrelations among these attributes, without considering their spatial arrangements in forests. Structural complexity includes structural components, the relationships among their attributes while considering their spatial arrangements in forests (Zenner 2004, McElhinny et al. 2005). Structural components can also be used to express ecosystem processes such as nutrient cycling (Spies 1998, McElhinny et al. 2005). For example, tree canopies, can influence local-climatic conditions in forests, and in turn affect other structural attributes, such as herbaceous plant diversity and their aboveground biomass production (Moore 2009). The distribution of individual structural attributes within and across forest ecosystems is driven by environmental conditions and anthropogenic disturbances (Varga et al. 2005).
The structural components of tropical forests and woodlands are experiencing high rates of degradation due to anthropogenic activities (Bunker et al. 2005, Strassburg et al. 2010), leading to losses of biodiversity (Sala et al. 2000, Barlow et al. 2007) and increase in atmospheric carbon emissions (Gibbs et al. 2007, Ciais et al. 2011). For example, African forest and woodland ecosystems varies from carbon sinks of about 3.2 Pg Cௗyr-1 to small sources (i.e.
from agriculture and other land use changes) of about 0.44 Pg Cௗyr-1 (Ciais et al. 2011). Forest and woodlands in Africa represents more than 30 % of the global forest cover (Malhi et al.
2013), and although woodlands and savannas account for lower carbon storage than forests, they cover an area three times larger than forests (Ciais et al. 2011). Woodlands and savannas in Africa accounts for about 65 % (range: 2.7-3.2 million km2) of the land-surface (Thomas and Packham 2007). However, over 40% of the growing human population in African countries rely on woodlands and savannas for their livelihoods (Mwampamba 2007, Chidumayo and Gumbo 2010, Bromhead 2012). Charcoal production and agricultural expansion are estimated to contribute about 20-25 % of woodland degradation in Africa (Chidumayo and Gumbo 2010).
It is projected that the demand for household fuel-wood in sub-Saharan countries may increase by 20 % from 2010 to 2030 (Bromhead 2012). In Tanzania, about 60-80 % of the energy used by the growing urban population are wood-based, such as charcoal, which are mostly supplied from miombo woodlands (Mwampamba 2007). Electricity covers only 30–40 % of the urban
2
energy supply (Kihwele et al. 2012). Thus, anthropogenic disturbances are the main drivers of changes in ecosystem structure, productivity and carbon balance in Africa (Ciais et al. 2011).
A diverse structure implies an increase in resource heterogeneity and is often associated with high biodiversity in forest and woodland ecosystems (McElhinny et al. 2005). Thus, forests and woodlands with a high plant diversity will use resources more efficiently and thereby may enhance the long-term carbon sequestration and storage, and nutrients cycling (Tilman 1997, Cardinale et al. 2007). Over the last few years, forest management has been geared towards accurate measurements, monitoring, reporting and verification of carbon stocks (DeFries et al. 2006). However, one of the challenges is to unveil the complex links between stand structural components and ecosystem functions (Naeem et al. 2009). This is because structural component assessment at stand scale can be used to inform management actions, such as harvesting or recreations i.e. scenic beauty (McElhinny et al. 2005). There has been a global demand for new sustainable ways to manage and finances ecosystem products and services. Thus, market based initiatives such the clean development mechanisms (CDM), reduced emissions from deforestation and forest degradation plus conservation and carbon stocks enhancement (REDD+), have emerged (Ebeling and Yasue 2008). There are hopes that if well-structured, these initiatives may yield tangible and sustainable benefits to local livelihoods from forest and woodland biodiversity and at same time positively affect global climate (Strassburg et al. 2010). There is also potentials to promote restorations of biological diversity in already degraded terrestrial ecosystems through the REDD+ initiatives if well implemented (Phelps et al. 2012).
One motivation for the REDD+ is to obtain accurate forest carbon stock data at minimum costs, for monitoring and decision making at local scale, to safeguard forest access rights, and to improve the involvement of local people in decision-making (Fry 2011, Skutsch 2012). However, most of the REDD+ readiness programs in developing countries (Danielsen et al. 2011), do not, or have little considerations of the entire structural components and their interactions with environmental conditions and anthropogenic disturbances. For example, REDD+ pilot studies in Tanzania were geared towards generating baselines data. However, in addition to financial and technical constrains (Burgess et al. 2010, Sills et al. 2013), there is a low ability to link carbon measurements to other forest structural components, their environmental conditions and anthropogenic disturbances. This has raised concern that there will be negative environmental consequences if the entire structural components are not well address in the REDD+ process (Dickson and Kapos 2012, Gardner et al. 2012, Phelps et al.
2012). It is crucial that accurate baseline data on structural components, such as species
3
diversity (Gardner et al. 2012) and their interactions with the environment, are part of the baselines for references emission levels, in order for REDD+ to be successful. Failure to account for various forest structural complexities and their interactions with physiographic conditions and anthropogenic disturbances may jeopardize future sustainable flow of goods and services. It also undermines the realization of the potential contributions of biodiversity in enhancing human wellbeing (Naeem et al. 2009).
The aim this study was to explore the existing relationships between tree species diversity, richness and evenness, aboveground carbon stocks and canopy foliage characteristics along gradients of physiographic conditions and anthropogenic disturbances in two vegetation types of Tanzania. The four major question addressed in this study were: (1) how do tree species richness relates to vertical heterogeneity, mean and depth specific soil nutrient availability? (2) do dominant tree species influence the richness, diversity, evenness and vertical structure heterogeneity of non-dominant tree species? (3) how do tree canopy characteristics relates to herbaceous biomass and tree species diversity? (4) how do the aboveground carbon stocks of trees relate to tree species richness, diversity and evenness along gradients of physiographic conditions and anthropogenic disturbances?
The specific objectives of this study were:
1. To explore the relationships between tree species richness and vertical heterogeneity, mean and depth specific soil nutrient availability in moist forest and miombo woodlands of Tanzania (Paper I).
2. To explore the relationships between the abundance of dominant tree species and richness, diversity, evenness, and vertical structure heterogeneity of non-dominant tree species in wet and dry miombo woodlands of Tanzania (Paper II).
3. To examine the relationships between canopy characteristics, herbaceous biomass, tree species diversity and environmental gradients in moist forest and miombo woodlands of Tanzania (Paper III).
4. To examine the relationships between aboveground carbon stocks of trees and tree species richness, diversity and evenness along gradients of physiographic conditions and anthropogenic disturbances in moist forest and miombo woodlands of Tanzania (Paper IV).
4
2.0 MATERIALS AND METHODS